The function of opioid receptor dimerization in ligand binding, selectivity, and signal transduction is investigated using structure-based modeling techniques. Pharmacological studies have indicated that opioid receptors form both homo- and heterodimeric complexes, providing new avenues for the design of ligands with modified pharmacology. A key step in understanding the function of dimerization in ligand binding and function involves the design and development of 3-dimensional models of the homo- and heterodimeric receptor structures. Initial models of mu-mu receptor homodimers and mu-delta receptor heterodimers will be constructed using concepts borrowed from chimeric studies of adrenergic and muscarinic receptors. Both experimental and theoretical studies have indicated those G protein-coupled receptors (GPCRs) form dimeric complexes interfaced at transmembrane (TM) helices V and VI, with a potential domain swapping mechanism that allows TM VI and VII to be translocated between monomeric units. Given the close sequence relationship of opioid, adrenergic, and muscarinic GPCRs, this structural motif will be applied to combine mu and delta receptor monomers into homo- and heterodimer complexes with TM V-VI interfaces in both contact and domain swapped configurations. The complexes will be evaluated using sequence analysis techniques, including evolutionary trace and correlated mutation methods, and further refined using a combination of molecular dynamics simulations and ligand docking studies. Based on the inherent asymmetry of the ligand-receptor complex and the predicted distances between the two pharmacophores of a bivalent ligand series synthesized in Project 1, a protocol is also developed to refine the global conformation of the dimer complex, allowing alternative TM interfaces and geometries to be explored. The approach takes advantage of the stereochemistry of the ligands to reduce the conformational space or size of the problem and conformational search techniques to pinpoint relevant bivalent ligands conformations for docking. Local refinements will be performed using molecular dynamics calculations employing a novel lipid solvation model to account for the effects of the membrane environment on dimer conformation and energetics. The results will be compared with the ligand binding data taken from Project 2 in an effort to explain the structural basis to receptor dimerization, ligand binding and selectivity.